Pascal Vicaire
Staff Software Engineer at Google
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolPascal Vicaire is a Staff Software Engineer in Seattle with nine years of professional experience building large-scale cloud and machine learning infrastructure at Google and AWS. He specializes in backend systems and distributed workflows using Java, C/C++, Ruby, SQL and a broad set of AWS services, and now focuses on BigQuery and multi-cloud data analytics at Google Cloud. Pascal has deep operational and DevOps chops—evidenced by contributions to Kubeflow Pipelines scheduling CRDs and enabling distributed TensorFlow training and model export for YouTube-8M—bridging infrastructure, APIs, and MLOps. His background in distributed wireless sensor research and a PhD-level computer science education inform a pragmatic approach to designing resilient, production-ready systems. Colleagues rely on him for solving hard scheduling, scale, and recovery problems that sit at the intersection of data analytics and ML platforms.
9 years of coding experience
15 years of employment as a software developer
Master's Degree, Computer Science, Master's Degree, Computer Science at University of Virginia
Github Skills (19)
kubernetes10
distributed-training10
resource-dictionary10
python10
kubeflow10
machine-learning10
scheduling10
go10
tensorflow10
pipeline10
kubernetes-pods10
workflow-management10
argo-workflows10
exporter9
apidoc9
Programming languages (5)
Github contributions (5)
kubeflow/pipelines
May 2018 - Apr 2019
Machine Learning Pipelines for Kubeflow
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:2 releases, 72 commits, 44 PRs in 10 months
Contributions summary:Pascal contributed to the development of a CRD (Custom Resource Definition) for scheduling Argo workflows. This included implementing features like cron schedules, periodic intervals, and start/end date configurations. The user was also involved in code generation processes and addressed code review comments. Their work focused on the infrastructure and API components related to scheduling and workflow management, aligning with DevOps and back-end development practices.
pipelinetektondata-sciencemachine-learningmlops
google/youtube-8m
Feb 2017 - Mar 2017
Starter code for working with the YouTube-8M dataset.
Role in this project:MLOps Engineer
MLOps EngineerContributions:19 commits, 17 PRs, 9 comments in 14 days
Contributions summary:Pascal contributed significantly to the training process within the YouTube-8M dataset project. Their work focused on enabling distributed TensorFlow training, integrating model export for batch prediction, and addressing bugs related to model recovery in distributed environments. This included modifying training scripts and implementing features to facilitate the exporting of trained models for later use, aligning with MLOps principles.
deep-learningdatasetmachine-learningyoutubeyoutube-8m
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.